Storing variations of data across different replication sites

Patent metadata

Field	Value
Publication number	US-11762816-B2
Application number	US-202117341202-A
Country	US
Kind code	B2
Filing date	Jun 7, 2021
Priority date	Jun 7, 2021
Publication date	Sep 19, 2023
Grant date	Sep 19, 2023

Abstract

A computer-implemented method according to one embodiment includes determining patterns of an application that utilizes a filesystem and/or properties of queries of the application. Data of the filesystem is stored across a plurality of replication sites of a data storage system. Based on the determined patterns of the application and/or the determined proper-ties of the queries of the application, a utility of storing at least some of the data of the filesystem in different variations at more than one of the replication sites is estimated. The estimated utility is compared against a predetermined utility threshold, and in response to a determination that the estimated utility is greater than the predetermined utility threshold, a write system call offered by the filesystem is modified to store the data in different variations at more than one of the replication sites.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
determining patterns of an application that utilizes a filesystem and/or properties of queries of the application, wherein data of the filesystem is stored across a plurality of replication sites of a data storage system;
based on the determined patterns of the application and/or the determined properties of the queries of the application, estimating a utility of storing at least some of the data of the filesystem, not already stored in different variations at more than one of the replication sites, in the different variations at more than one of the replication sites;
comparing the estimated utility against a predetermined utility threshold; and
in response to a determination that the estimated utility is greater than the predetermined utility threshold, modifying a write system call offered by the filesystem to store the at least some of the data in at least some of the different variations at more than one of the replication sites.

2. The computer-implemented method of claim 1 , comprising:
storing the at least some of the data of the filesystem in the at least some of the different variations at the more than one of the replication sites,
wherein each of the at least some of the different variations of the data is stored with differentiating metadata to be used for reconstructing the variation of the data stored at another one of the replication sites, wherein the differentiating metadata stored with a first variation of the data stored on a first of the replication sites identifies differences between the first variation of the data and a second variation of the data stored on a second of the replication sites;
outputting a scheduler interface to the application, wherein the scheduler interface is configured to hint a location to read data for a set of the queries of the application; and
instructing the scheduler interface to be used for sending a read system call to one of the replication sites having a variation of the data stored thereon that fulfills the read system call.

3. The computer-implemented method of claim 1 , wherein estimating the utility of storing at least some of the data of the filesystem in the different variations at more than one of the replication sites includes: determining whether the data fulfills a first set of the queries while the data is in a first variation stored on a first of the replication sites and fulfills a second set of the queries while the data is in a second variation stored on a second of the replication sites, wherein the estimated utility is determined to be greater than the predetermined utility threshold in response to a determination that the data fulfills the first set of the queries while the data is in the first variation and fulfills the second set of the queries while the data is in the second variation, wherein the data in the first variation is stored in a different parseable format than a parseable format that the data in the second variation is stored in.

4. The computer-implemented method of claim 1 , wherein the estimated utility is based on time complexities associated with reading and writing the at least some of the data of the filesystem in the different variations at more than one of the replication sites and storage complexities associated with storing the at least some of the data of the filesystem in the different variations at more than one of the replication sites.

5. The computer-implemented method of claim 1 , wherein the determined patterns of the application are selected from the group consisting of: determined characteristics of the application, a determined input/output (I/O) nature of the application and determined replication properties of the filesystem, and estimated times that it takes for the application to access the data from the different replication sites.

6. The computer-implemented method of claim 1 , wherein determining the properties of the queries of the application includes: determining a minimum amount of the data an answer to at least one of the queries includes, and determining an average time that responding to the queries consumes.

7. The computer-implemented method of claim 1 , comprising: in response to a determination that the estimated utility is less than or equal to the predetermined utility threshold, storing copies of the at least some data at more than one of the replication sites.

8. A computer program product, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions readable and/or executable by a computer to cause the computer to:
determine, by the computer, patterns of an application that utilizes a filesystem and/or properties of queries of the application, wherein data of the filesystem is stored across a plurality of replication sites of a data storage system;
based on the determined patterns of the application and/or the determined properties of the queries of the application, estimate, by the computer, a utility of storing at least some of the data of the filesystem, not already stored in different variations at more than one of the replication sites, in the different variations at more than one of the replication sites;
compare, by the computer, the estimated utility against a predetermined utility threshold to determine whether to modify a write system call offered by the filesystem to store the at least some of the data of the filesystem in different variations at more than one of the replication sites; and
in response to a determination that the estimated utility is greater than the predetermined utility threshold, modify, by the computer, the write system call offered by the filesystem to store the at least some of the data in the at least some of the different variations at more than one of the replication sites.

9. The computer program product of claim 8 , the program instructions readable and/or executable by the computer to cause the computer to:
store, by the computer, the at least some of the data of the filesystem in the at least some of the different variations at the more than one of the replication sites,
wherein each of the at least some of the different variations of the data is stored with differentiating metadata to be used for reconstructing the variation of the data stored at another one of the replication sites;
output, by the computer, a scheduler interface to the application, wherein the scheduler interface is configured to hint a location to read data for a set of the queries of the application; and
instruct, by the computer, the scheduler interface to be used for sending a read system call to one of the replication sites having a variation of the data stored thereon that fulfills the read system call.

10. The computer program product of claim 8 , wherein estimating the utility of storing at least some of the data of the filesystem in the different variations at more than one of the replication sites includes: determining whether the data fulfills a first set of the queries while the data is in a first variation stored on a first of the replication sites and fulfills a second set of the queries while the data is in a second variation stored on a second of the replication sites, wherein the estimated utility is determined to be greater than the predetermined utility threshold in response to a determination that the data fulfills the first set of the queries while the data is in the first variation and fulfills the second set of the queries while the data is in the second variation.

11. The computer program product of claim 8 , wherein the estimated utility is based on time complexities associated wit

Assignees

• IBM

Inventors

• Thorat Pushkaraj Balasaheb
• Naik Sandeep
• Eda Sasikanth
• Shingornikar Shashank
• Ukirde Digvijay

Classifications

• G06F16/164 Primary

  File meta data generation · CPC title

• G06F16/1844 Primary

  Management specifically adapted to replicated file systems · CPC title

• G06F3/065

  Replication mechanisms · CPC title

• G06F3/067

  Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS] · CPC title

• G06F3/0604

  Improving or facilitating administration, e.g. storage management · CPC title